A brief history of all natural products

Bioactive natural products have been used for millennia, with examples dating back to the dawn of recorded history. Among of the earliest sources of natural products are cedar oil (Cedrus sp.), cypress oil (Cupressus sempevirens), licorice oil (Glycyrrhiza glabra), myrrh (Commiphora sp.) and poppy juice (Papaver somniferum) [1]. One of the first recorded chemists, Tapputi-Belate- kallim, used such materials to prepare perfumes while serving the Royal Palace in Mesopotamia [2]. Early natural products also include an anti- biotic material from the moldy curd of soybean, used to fight infections by Chinese populations, as well as tetracycline antibiotics produced by Streptomyces, present in beer consumed by the early Sudanese–Nubian civilization [3,4].

Certainly, present day FDA-approved natural pro- ducts are better characterized, although some remain as complex mixtures, including Veratrum viride root, ergoloid mesylates and sinecate- chins. As part of a program to isolate and study novel natural products, the Yale Center for Molecular Discovery evaluated the history of all FDA-ap- proved new molecular entities (NMEs) derived from natural products. The analgesic properties of opium had been well-known throughout history when Friedrich Serturner isolated the active component, morphine, from opium resin in 1804. First distributed by Serturner in 1817, Heinrich Merck initiated commercialization of the plant alkaloid in 1827, more than a century before the founding of the modern FDA [5]. Like morphine, other natural products that are still in use today were introduced long before they garnered official regulatory approval.

For in- stance, guaifenesin is a semisynthetic plant de- rivative from guaiacum resin, which was used during the 16th century to (ineffectively) treat syphilis, long before guaifenesin gained its ulti- mate approval as a cough suppressant in 1952 [6]. Likewise, the therapeutic properties of the foxglove plant, Digitalis lanata, were utilized as early as the 16th century. However, the active component, digoxin, was not isolated until the 1930s [7]. Presently, the definition of what constitutes a natural product varies across the scientific and medicinal communities. For therapeutic natural products, acceptable definitions broadly include:

• (i) unregulated organisms or natural materials;
• (ii) FDA-regulated, unmodified natural materials or compounds;
• (iii) a naturally occurring compound that has been chemically modified (also called a semisynthetic); and
• (iv) a purely synthetic me- dicinal compound inspired by a natural compound [8,9].

For the purposes of this study, we define a natural product as any FDA-approved, unmodified natural material or compound semisynthetic derivatives, or synthetic structures which were conceptually derived from a natural product.

For the purposes of this review, we differentiated natural products into two broad categories: (i) those found in mammalian organ- isms, including metabolites; and (ii) those isolated from plants, bacteria, fungi, marine organisms and non-mammalian animal species. Thus, we refer- ence the mammalian compounds as biochemical natural products (BCNPs) to distinguish them from non-mammalian natural products (NMNPs). We have further excluded mammalian protein- based biologics from our BCNP definition and analysis.

We felt it important to differentiate the impact of natural product research from biolo- gics-based research in drug discovery, which was the subject of an earlier review by our group [10]. By the end of 2013, the FDA had approved 547 natural products and derivatives. Including their derivatives, NMNPs and BCNPs represent more than one-third (38%) of all FDA-approved NMEs.

The absolute rate of introduction for FDA-ap- proved natural product NMEs rose steadily since the 1930s, peaking in the 1990s at an average annual approval rate of 10.3 NMEs (Fig. 1a).

Although impressive, this growth must be con- sidered in the context of overall increases in FDA approvals (i.e. of synthetics and biologics), which likewise increased over the same time period. Indeed, the relative number of natural-product- based approvals varied from as low as one-third (33% in the 1950s) to more than half (in the 1930s and 1970s) of all approvals. Since the 1970s, the relative and then absolute number of natural-product-based NMEs began to decline and today stands at fewer than one-quarter (24%) or an average of 7.7 natural product NMEs per year. Having established the overall trends in ap- proval rates, we began to assess the sources and types of these natural products.

For example, the relative prominence of unmodified NMNPs (neither biochemical nor derivatives) as well as unmodified BCNPs peaked in the 1930s and generally declined over time. Excluding their derivatives, just 9.8% of all FDA-approved NMEs are unmodified NMNPs or BCNPs, and each of these led to an average of three additional semisynthetic or synthetic derivatives. Over time, the unmodified NMNPs and BCNPs have largely been replaced by their derivatives which, through modification, often have increased ef- ficacy, safety and/or delivery relative to the parent compound.

Analysis of unmodified products and their derivatives

The pursuit of new natural products requires intensive investments in time and resources; in part because of the effort required to isolate and identify these substances. For these reasons and others, such as the difficulty of total synthesis, the pharmaceutical industry has progressively moved away from natural products and toward synthetic compounds [9]. This change appears to explain a notable trend in our data, which emerged by separating the unmodified natural products from their semisynthetic and synthetic derivatives (Fig. 1b).

Over time, semisynthetic and synthetic derivatives of natural products have become increasingly predominant. Un- modified products and their derivatives repre- sented 43% and 14%, respectively, of NME approvals before 1940. Since then, unmodified products dropped to 9.5% of all approved NMEs, whereas their derivatives doubled to 28%. These trends continue today.

Since the year 2000, unmodified natural products and their deriva- tives represent 5.3% and 28%, respectively. A landscape assessment revealed a total of 307 NMNPs, or just over one-fifth (21%) of all approved NMEs, by the end of 2013; proving further that this subset encompasses unmodi- fied NMNPs (7.5%), semisynthetic derivatives (9.6%) and synthetic derivatives (4.0%).

When viewed over time, the absolute rate of unmodified NMNPs peaked during the 1960s with an average annual rate of 2.1 NMEs per year, with a yearly high of seven NMEs approved in 1963 (Fig. 1b). As emphasis on unmodified NMNPs waned, their semisynthetic and synthetic deri- vatives rose, peaking through the 1980s and 1990s with an average annual rate of 4.2 NMEs per year, with a yearly high of ten NMEs approved in 1995.

A parallel assessment revealed a total of 240 mammalian-derived BCNPs (or a total of 16.5% of all FDA-approved NMEs) by the end of 2013. Using a similar approach as described above, we compared parental versus modified variants and found that unmodified BCNPs captured 33 NMEs (2.3% of total NMEs) with their semisynthetic derivatives and synthetic derivatives encompassing 1.2% and 13% of all FDA-approved NMEs, respectively. 

When viewed over time, the absolute number of approvals of unmodified BCNPs reached a peak in the 1980s with an average annual rate of 0.7 NMEs per year and with a yearly high of three NMEs approved in 1986. Their semisynthetic and synthetic deriva- tives have gained in popularity over time, peaking in the current decade at an average annual rate of five new approvals per year, with a yearly high of six NMEs in the year 2011.

Analysis of natural products from plants, microbes and other sources

An assessment of the source of natural products revealed that almost half (44%) were derived from mammals (biochemical natural products) with the primary species of origin being (in order frequency) bovine, porcine, equine, canine and human. The remaining NMNP sources were dominated by plants (25% overall; 45% NMNPs), bacteria (16% 29% NMNPs) fungi (12% 22% (Fig. 2a). first plant NME was morphine, approved in 1827, before creation modern FDA 1938. bacterial fungal NMEs 1946 (streptomycin) 1953 (phenoxymethyl penicillin).

Plant products represented more than one- fifth (22%) all 1950, declining 50% to 8.7% since that time. By contrast, microbial derivatives encom- passed 3.9% growing thereafter 11%. Upon close analysis, microbial-derived nearly equally (6.2%) (4.9%).

Despite fact overall rate natural product approvals has been declining, continue cap- ture 5.6% 8.9%, respectively, 2000. Among sources, a few genera are heavily terms FDA-approved NMEs, including Streptomyces, Penicillium, Cephalosporium, Salix, Guaiacum Digitalis. These highly biased toward anti- infective anticancer agents 3a). Spe- cifically, total 223 antibiotics could be broadly distinguished based on their abilities target, for example, (151 NMEs; 49% pathogenic (seven 2.3% or human cancer cells (40 13% NMNPs). 

Looking closely at antibacterial agents, it is notable 69% originate from products, 97% iso- lated derived microbes This tendency remains true today, even amid current decline absolute relative rates introduction NMEs. Since year 2000, 77% 100% which microbes. mechanistic actions targets varied widely but certain subsets predominated. Antibacterial tended target topoisomerases (tetra- cyclines), peptidyltransferases (macrolides ascamycins), ribosomes (aminoglycosides) transpeptidases (beta-lactams).

Natural-product-based have also used many pathways indications. A large number G-protein-coupled receptors (GPCRs; Fig. 3b).

At least 35 different natural-product-based opioid neurotransmitter receptors, most (94%) plants. myriad statin-based drugs HMG- CoA reductase hypercholesterolemia [11] Penicillium molds, sal- icylic-acid-based medicines cycloox- ygenases [12] originally isolated willow trees (Supplementary Table S1).

Concluding remarks and implications

Historically, natural products their deriva- tives played a significant part in medicine, pro- ducing many of the early NMEs nearly half all approved NMEs. However, during final decades 20th century, were de-prioritized by research organizations, seemingly favor workflows that more di- rectly supported high-throughput screening formats [13].

An increasing number reports are drawing attention to importance sustained drug discovery pipeline for unmodified products, which would facilitate chemical diversity provide several each NME discovered [14,15]. Some these assessed past 30 years but, our knowledge, present report is first offer landscape view FDA-approved product NMEs, providing trends since century highlighting more-detailed specifics about origins every NME.

In another report, we recently noted organizations leaving field antibacterial discovery, even though antimi- crobial resistance on rise [16]. Given substantial fraction antibiotics originate from microbial sources, it seems clear attempts bridge antibiotic gap will need involve an effort designed investigate fungi bacteria. As previously discussed, there be order reach success, could governmental and/or non-govern- mental support lay foundation insure long-term success.

Acknowledgments

This work was conducted as part of a project at the Yale Center for Molecular Discovery to develop collection all FDA-approved small molecules resource screening emphasize drug repurposing. Please contact authors if you or your organization would be interested in potential participation this project.

Appendix A. Supplementary data material related article can found, online version  http://dx.doi.org/10.1016/j.drudis.2015.01.009

References

1. 1 Cragg, G.M. and Newman, D.J. (2001) Natural product drug discovery in the next millennium. Pharm. Biol. 39 (Suppl. 1), 8–17
2. 2 Levey, M. (1956) Babylonian chemistry: a study of Arabic and second millenium B.C. perfumery. Osiris 12, 376–389
3. 3 Moellering, R.C., Jr (1995) Past, present, and future of antimicrobial agents. Am. J. Med. 99, 11s–18s
4. 4 Armelagos, G.J. et al. (2001) Tetracycline consumption in prehistory. In Tetracyclines in Biology, Chemistry and Medicine (Nelson, M. et al. eds), pp. 219–236, Springer
5. 5 Schiff, P.L. (2002) Opium and its alkaloids. Am. J. Pharm. Educ. 66, 188–196
6. 6 Frith, J. (2012) Syphilis – its early history and treatment until penicillin, and the debate on its origins. J. Mil. Veterans Health 20, 4
7. 7 Gerwick, W.H. et al. (2007) Plant sources of drugs and chemicals. In Encyclopedia of Biodiversity.
8. 8 Schmidt, B. et al. (2008) A natural history of botanical therapeutics. Metabolism 57 (Suppl. 1), 3–9
9. 9 Koehn, F.E. and Carter, G.T. (2005) The evolving role of natural products in drug discovery. Nat. Rev. Drug Discov. 4, 206–220
10. 10 Kinch, M.S. (2014) An overview of FDA-approved biologics medicines. Drug Discov. Today http:// dx.doi.org/10.1016/j.drudis.2014.09.003
11. 11 Endo, A. (1992) The discovery and development of HMG- CoA reductase inhibitors. J. Lipid Res. 33, 1569–1582
12. 12 Mahdi, J. et al. (2006) The historical analysis of aspirin discovery, its relation to the willow tree and antiproliferative and anticancer potential. Cell Prolif. 39, 147–155
13. 13 Roggo, S. et al. (2007) Natural products in drug discovery. Chimia 61, 312
14. 14 Newman, D.J. and Cragg, G.M. (2012) Natural products as sources of new drugs over the 30 years from 1981 to 2010. J. Nat. Prod. 75, 311–335
15. 15 Gerwick, W.H. and Moore, B.S. (2012) Lessons from the past and charting the future of marine natural products drug discovery and chemical biology. Chem. Biol. 19, 85–98
16. 16 Kinch, M.S. et al. (2014) An analysis of FDA-approved drugs for infectious disease: antibacterial agents. Drug Discov. Today 19, 1283–1287

• Eric Patridge
•  Peter Gareiss
• Michael S. Kinch
• Denton Hoyer

1. 1  Yale Center for Molecular Discovery, West Haven , CT 06516, USA
2. 2 Washington University in St Louis, St Louis, MO 63110, USA